How to build a reinforcement learning environment for a DCDC converter?

I will use reinforcement learning algorithm to replace the PI controller to achiveve close loop control for my dual active bridge DCDC converter. Should I build my own customed environment in simulink? Is there any example for me to follow?

Answer 1

[Emmanouil Tzorakoleftherakis]

I would look at this example which starts with a model that has a PID controller and shows how to replace it with an RL Agent.